Content uploaded by Don Davis
Author content
All content in this area was uploaded by Don Davis on Jul 28, 2015
Content may be subject to copyright.
PLEASE SCROLL DOWN FOR ARTICLE
7KLVDUWLFOHZDVGRZQORDGHGE\
>9LUJLQLD&RPPRQZHDOWK8QLYHUVLW\@
2Q
-XQH
$FFHVVGHWDLOV
$FFHVV'HWDLOV>VXEVFULSWLRQQXPEHU@
3XEOLVKHU
5RXWOHGJH
,QIRUPD/WG5HJLVWHUHGLQ(QJODQGDQG:DOHV5HJLVWHUHG1XPEHU5HJLVWHUHGRIILFH0RUWLPHU+RXVH
0RUWLPHU6WUHHW/RQGRQ:7-+8.
-RXUQDORI3HUVRQDOLW\$VVHVVPHQW
3XEOLFDWLRQGHWDLOVLQFOXGLQJLQVWUXFWLRQVIRUDXWKRUVDQGVXEVFULSWLRQLQIRUPDWLRQ
KWWSZZZLQIRUPDZRUOGFRPVPSSWLWOHaFRQWHQW W
5HODWLRQDO+XPLOLW\&RQFHSWXDOL]LQJDQG0HDVXULQJ+XPLOLW\DVD
3HUVRQDOLW\-XGJPHQW
'RQ('DYLVD-RVKXD1+RRNE(YHUHWW/:RUWKLQJWRQ-UD'DU\O59DQ7RQJHUHQD$XEUH\/
*DUWQHUD'DYLG--HQQLQJV,,D5REHUW$(PPRQVF
D'HSDUWPHQWRI3V\FKRORJ\9LUJLQLD&RPPRQZHDOWK8QLYHUVLW\E'HSDUWPHQWRI3V\FKRORJ\
8QLYHUVLW\RI1RUWK7H[DVF'HSDUWPHQWRI3V\FKRORJ\8QLYHUVLW\RI&DOLIRUQLD'DYLV
2QOLQHSXEOLFDWLRQGDWH$SULO
7RFLWHWKLV$UWLFOH'DYLV'RQ(+RRN-RVKXD1:RUWKLQJWRQ-U(YHUHWW/9DQ7RQJHUHQ'DU\O5*DUWQHU
$XEUH\/-HQQLQJV,,'DYLG-DQG(PPRQV5REHUW$5HODWLRQDO+XPLOLW\&RQFHSWXDOL]LQJDQG0HDVXULQJ
+XPLOLW\DVD3HUVRQDOLW\-XGJPHQW-RXUQDORI3HUVRQDOLW\$VVHVVPHQW٢
7ROLQNWRWKLV$UWLFOH'2,
85/KWWSG[GRLRUJ
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
Journal of Personality Assessment,93(3), 225–234, 2011
Copyright C
!Taylor & Francis Group, LLC
ISSN: 0022-3891 print / 1532-7752 online
DOI: 10.1080/00223891.2011.558871
Relational Humility: Conceptualizing and Measuring Humility as
a Personality Judgment
DON E. DAV I S ,1JOSHUA N. HOOK,2EVERETT L. WORTHINGTON JR.,1DARYL R. VAN TONGEREN,1AUBREY L. GARTNER,1
DAV I D J. JENNINGS II,1AND ROBERT A. EMMONS3
1Department of Psychology, Virginia Commonwealth University
2Department of Psychology, University of North Texas
3Department of Psychology, University of California, Davis
The study of humility has progressed slowly due to measurement problems. We describe a model of relational humility that conceptualizes
humility as a personality judgment. In this set of 5 studies, we developed the 16-item Relational Humility Scale (RHS) and offered initial evidence
for the theoretical model. In Study 1 (N=300), we developed the RHS and its subscales—Global Humility, Superiority, and Accurate View of
Self. In Study 2, we confirmed the factor structure of the scale in an independent sample (N=196). In Study 3, we provided initial evidence
supporting construct validity using an experimental design (N=200). In Study 4 (N=150), we provided additional evidence of construct validity
by examining the relationships between humility and empathy, forgiveness, and other virtues. In Study 5 (N=163), we adduced evidence of
discriminant and incremental validity of the RHS compared with the Honesty-Humility subscale of the HEXACO–PI (Lee & Ashton, 2004).
With the rise of positive psychology, the study of virtue is
thriving (Lopez & Snyder, 2009). Literatures on virtues such
as forgiveness (Worthington, 2005), gratitude (Emmons & Mc-
Cullough, 2004), hope (Snyder, Rand, & Sigmon, 2002), and
optimism (Carver & Scheier, 1994) have expanded rapidly. In-
deed, even the study of modesty is flourishing (Sedikides, Gregg,
&Hart,2007).Instarkcontrast,thestudyofhumilityhasbeen
stagnant, mainly due to measurement problems (Tangney, 2009).
One reason that modesty has been studied more rigorously
than humility is that definitions of modesty have led to generally
accepted measurement strategies. Namely, two types of mod-
esty have been defined (Gregg, Hart, Sedikedes, & Kumashiro,
2008). Intrapersonal modesty refers to having an accurate view
of self—not too high or low. This has been measured as self-
enhancement, which is the degree to which a person has an
overly positive view of self. Interpersonal modesty refers to the
tendency to moderate praise or recognition in socially acceptable
ways, particularly in public settings. This has been measured by
observing how participants act in experimental studies. For ex-
ample, Tice, Butler, Muraven, and Stillwell (1995) found that
people tend to portray themselves more modestly with friends
than strangers. These two modesty constructs might not neces-
sarily overlap.
In contrast, humility has been more difficult to define. Def-
initions of humility include, but are not limited to, both types
of modesty. Humility also includes other interpersonal qual-
ities, such as respect and empathy during conflict, openness
toward different cultures or worldviews, and acceptance of self
as subordinate to God or the transcendent (for a review of def-
initions, see Davis, Worthington, & Hook, 2010). Furthermore,
humility involves thinking and acting in these ways for the right
reasons (i.e., other-oriented motivations). It involves integrity
Received April 26, 2010; Revised September 8, 2010.
Address correspondence to Don E. Davis, Department of Psychology, Vir-
ginia Commonwealth University, 806 West Franklin Street, Richmond, VA
23284–2018; Email: don.davis@aya.yale.edu
of belief, behavior, and motivation (Davis, Worthington, et al.,
2010; Tangney, 2009). Such multifaceted definitions have made
humility difficult to measure.
Given that humility has been defined in such aspirational
terms, humility researchers have doubted the validity of self-
reports of humility. That is, people who claim to be very humble
would seem to be bragging about their humility, something truly
humble people would not do. People tend to self-enhance on val-
ued traits such as humility (John & Robins, 1993). Researchers
have hypothesized that to the degree that people are truly hum-
ble, they will more modestly report their own humility on self-
report measures (Davis, Worthington, et al., 2010). Therefore,
truly humble people might modestly underreport their humility,
moderately humble people might overestimate their humility
some, and people low in humility (e.g., narcissists) might over-
estimate their humility agreatdeal.Although this modesty
effect for self-reports of humility has not been tested empiri-
cally, humility researchers have assumed that it undermines the
validity of self-reports. This problem led Tangney (2005) to con-
clude that humility might be “one construct that is simply not
amenable to self-reports” (p. 415). Given the problems with self-
reports, researchers have sought other strategies of measuring
humility.
FOUR APPROACHES TO MEASURING HUMILITY
Davis, Worthington, et al. (2010) reviewed and critiqued four
approaches to measuring humility: self-reports, social compar-
isons, implicit measures, and informant ratings. We briefly de-
scribe each approach and its limitations.
Self-Reports
Despite the strong warnings of humility researchers, self-
report measures of humility have been created. Arguably,
the strongest published measure of humility is the Honesty-
Humility (HH) subscale of the HEXACO–PI (Lee & Ashton,
2004). The HH was derived through factor analysis. Whereas
225
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
226 DAVIS ET AL.
prior lexical studies of personality inventories have generally
found five personality factors (Costa & McCrae, 1992), Lee and
Ashton extracted this sixth factor, which is most closely aligned
with Agreeableness from the traditional five-factor model. The
HH has four subscales: Fairness, Sincerity, Greed-Avoidance,
and Modesty. With the addition of the HH, the HEXACO–PI
has demonstrated evidence of incremental predictive validity for
constructs related to low sociability, such as psychopathy, lack
of sexual restraint, Machiavellianism, boredom proneness, and
social adroitness (for a review, see Ashton & Lee, 2007). The
HH subscale has made a substantial contribution to the person-
ality literature. However, a limitation of the scale is that it only
aligns with one aspect of humility—namely, modesty as a lack
of superiority (e.g., I am an ordinary person who is no better
than others; I wouldn’t want people to treat me as though I were
superior to them). The other subscales of the HH (Sincerity,
Fairness, and Greed-Avoidance) do not align as well with how
humility has been defined.
Social Comparisons
Rowatt and his colleagues adapted a social-comparison
method from the self-enhancement literature to assess humil-
ity. They defined humility as having an accurate view of self
(Rowatt, Ottenbreit, Nesselroade, & Cunningham, 2002). Par-
ticipants compared themselves to others (e.g., the average col-
lege student) on how well they adhered to 12 biblical command-
ments. One limitation of this method is that it cannot discern
whether scores reflect attitudes of superiority or accurate per-
ceptions of true differences. For example, Rowatt et al. (2002)
found that people with high intrinsic religiosity viewed them-
selves as better at adhering to biblical commandments than oth-
ers. Perhaps people who were intrinsically religious (i.e., reli-
gious for its own sake rather than for extrinsic reasons, such as
social contact or prestige) actually were more likely to adhere
to religious teachings than those who were not as intrinsically
religious. Thus, an individual’s actual trait was not estimated
or accounted for, which is a confound that threatens the inter-
nal validity of this approach (for a discussion, see Kwan, John,
Kenny, Bond, & Robins, 2004; Kwan, John, Robins, & Kuang,
2008).
Implicit Measures
Rowatt and other colleagues also developed the Implicit As-
sociations Test of Humility versus Arrogance (IAT–HA; Powers,
Nam, Rowatt, & Hill, 2007; Rowatt et al., 2006). Implicit mea-
sures have been used to study other constructs that are prone to
impression management or socially desirable responding (e.g.,
attitudes of racial prejudice or self-esteem; Greenwald & Ba-
naji, 1995). For the IAT–HA, a computer program flashes words
with either a humble or arrogant connotation on the monitor. Par-
ticipants pair each word with self or other by pressing a corre-
sponding key (e.g., S for self, O for other) as quickly as possible.
Participants’ reaction times are used to infer their level of hu-
mility. The logic behind the IAT is that people should perform
cognitively similar pairings more quickly than dissimilar pair-
ings. For example, a humble person should pair humble words
with self more quickly than arrogant words. The approach is
apromisinginnovationinthemeasurementofhumility.How-
ever, the IAT–HA needs additional evidence of construct and
criterion-related validity. It was found to correlate with self-
report measures of low humility (e.g., narcissism or an author-
created scale that contrasted humility and arrogance words), but
not with established self-report or other-report measures of hu-
mility or modesty (Powers et al., 2007; Rowatt et al., 2006).
Given that the IAT–HA was designed to address problems with
self-reports, it is especially important to provide evidence of
criterion validity, showing that the measure can predict humble
behaviors (e.g., doing charitable work without apparent recog-
nition or other ulterior motives) or whether one is viewed as
humble by others.
Informant Ratings
Finally, researchers have used informant ratings to measure
humility. Namely, items from self-report measures of humility
were adapted for use as other-reports. The approach has gen-
erally been used as evidence of criterion-related validity for
another measurement strategy (e.g., self-report). Some studies
have failed to find a relationship between informant ratings
and the humility measure being studied (e.g., self-report, Park
&Peterson,2006;IAT–HA,Rowattetal.,2006).Otherstud-
ies have found moderate agreement. For example, de Vries,
Lee, and Ashton (2008) reported that self–other agreement on
the HEXACO–HH was as high as .60 by romantic partners,
but ratings by casual acquaintances, coworkers, or friends were
substantially lower (r=.22, .28, and .30, respectively).
One advantage of using other-reports to study humility is that
it avoids the paradox of having people self-report their own
humility. Whereas claiming to be humble might be immodest,
describing someone else as humble is not. Thus, a straightfor-
ward measure of humility judgments, completed by informants,
might be used to assess humility, at least within the context of a
specific relationship.
Still, the approach has several limitations. First, as we dis-
cussed earlier, existing measures of humility do not align fully
with how humility has been defined. Second, using informants
can be a time-intensive and expensive method of research. The
Internet, however, has greatly increased the feasibility of such
research (Vazire, 2006). Third, until recently, researchers had
not theoretically elaborated on what it might mean to use other-
reports as a primary method of studying humility.
AMODEL OF RELATIONAL HUMILITY
To address the need for theoretical elaboration, Dav i s , Wo r -
thington, et al. (2010) proposed a model of relational humility.
The model aligns the study of humility with a large literature on
personality judgements (for reviews, see Funder, 1995; Kenny,
2004; Kwan et al., 2004; McCrae & Weiss, 2007).
Relational humility was defined as an observer’s judgment
that a target person (a) is interpersonally other-oriented rather
than self-focused, marked by a lack of superiority; and (b) has
an accurate view of self—not too inflated or too low. Drawing
on recent theorizing on moral emotions (Tangney, Stuewig, &
Mashek, 2007), we suggested that people might judge humil-
ity by focusing on the target person’s ability to cultivate pos-
itive, other-oriented emotions, as well as regulate self-focused
emotions in modest and socially acceptable ways. Furthermore,
an observer might infer someone’s view of self from his or
her behavior, such as facial expressions, language, or abil-
ity to learn from mistakes. For example, people who consis-
tently overcommit and thus fail to meet their obligations might
be viewed as having an inflated view of self. Accordingly,
relational humility is measured by having an informant, who is
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
RELATIONAL HUMILITY 227
in a relationship with a person, rate that person’s humility. This
subjective judgment—including distortions caused by relation-
ship factors, such as characteristics of the judge, the target, or the
information available to the judge (Funder, 1995)—is the con-
struct of interest.
Davis, Worthington, et al. (2010) also proposed that judg-
ments of humility generally lead people to deepen their rela-
tionship with a target person. According to a theory of emotion
by Baumeister, Vohs, DeWall, and Zhang (2007), conscious
emotions do not cause behavior; rather, they cause a person
to think about what just happened. Specifically, negative emo-
tions cause people to examine details to prevent an undesired
event, whereas positive emotions provide a global stamp of ap-
proval, allowing a person to turn his or her attention toward the
future (Fredrickson, 1998). When a judge perceives positive,
other-oriented emotions (e.g., empathy, sympathy, or gratitude)
being expressed, the judge might interpret this as a sign of
other-oriented respect and value. Accordingly, seeing the tar-
get person as other-oriented should evoke positive emotions
that cause the judge to view the target person as more humble.
In contrast, when a judge perceives self-focused emotions (e.g.,
shame, pride, or contempt) being expressed in socially offensive
ways, this could signal selfishness and disrespect, which should
evoke anxiety in the judge (Tangney et al., 2007), causing the
judge to view the target person as less humble.
To accurately judge humility, a person must actually ob-
serve behaviors that are relevant to humility (Funder, 1995).
We posit that humility is best observe d i n t h r e e s i t u a t i o n s t h a t
make being other-oriented difficult: (a) honor or recognition,
(b) hierarchical roles, and (c) conflict. Each of these situations
tends to threaten the hierarchy of relationships. For example,
when a person receives an honor or recognition, other people
might feel jealous or question whether their relationship with
the person being honored will change. In hierarchical roles (e.g.,
parent–child, manager–subordinate, or teacher–student), subor-
dinates who want greater status or fear exploitation might rebel
against the leader’s authority. During conflict, people might ex-
perience negative emotions (e.g., disgust or contempt) that lead
them to feel morally superior to others, making empathy and re-
spect difficult. Thus, in all three situations, humility counteracts
the naturally tendency of these situations to cause instability in
relationships.
PRESENT STUDIES
The purposes of this set of studies were to (a) provide initial
evidence for the model of relational humility, (b) develop a
face-valid measure of humility judgments that aligned with our
definition of relational humility, and (c) provide initial evidence
that both the HH and the Relational Humility Scale (RHS) can
be used to assess judgments of humility.
We created the 71-item RHS–71 by dev e l o p i n g i t e m s t o a s -
sess a range of qualities that have been theorized to be part of
humility, including modesty, lack of superiority, accurate view
of self, respect and value for others, as well as a number of items
that directly assess humility (e.g., He or she is the most humble
person I know).
In Study 1, we administered the RHS–71 to a sample of un-
dergraduate students and used exploratory factor analysis (EFA)
to examine the structure of the scale and evaluate items. The fi-
nal version yielded a three-factor scale with 16 items, called the
RHS. In Study 2, the factor structure of the RHS was replicated
on an independent sample using confirmatory factor analysis
(CFA). In Study 3, we conducted an experiment. Undergraduate
participants nominated the most or least humble person they
have known and then rated that person’s degree of humility. In
Study 4, we examined evidence for the construct- and criterion-
related validity of the RHS. We also examined evidence for our
model of relational humility. Namely, we explored how judg-
ments of humility affect relationships undergoing conflict (i.e.,
one of three situations in which humility is most readily observ-
able). In Study 5, we examined evidence of discriminant and
incremental validity, comparing the RHS to the HH subscale of
the HEXACO–PI (Lee & Ashton, 2004). As in Study 4, Study 5
examined how humility judgments were related to forgiveness
of an offense. We used both the RHS and the HH as measures
of relational humility.
STUDY 1
The purposes of Study 1 were to (a) use EFA to determine
the factor structure of the RHS–71; (b) winnow items to create a
brief, face-valid measure of humility judgments; and (c) provide
initial evidence of estimated internal consistency of the scale and
subscales.
Method
Participant s . Participants were 300 undergraduate students
(166 female) from a large urban university. Ages ranged from
18 to 34 (M=19.1, SD =2.49). The sample was ethnically
diverse (57.3% White/Caucasian, 15.3% Black/African Amer-
ican, 18.0% Asian/Asian American, 4.0% Latino/Latina, and
4.7% other or did not report).
Measure: Humility Judgments (RHS–71). We generated
a list of 71 face-valid items—based on various definitions of
humility—that corresponded with our theoretical conceptual-
ization of relational humility. Participants rated the humility of a
parent by indicating their agreement with items using a 5-point
rating scale ranging from 1 (strongly disagree)to5(strongly
agree). Example items include “He/she has a humble charac-
ter,” “He/she has a big ego,” and “He/she knows him/herself
well.”
Procedure. Participants were recruited from undergraduate
classes and participated in exchange for partial course credit. Af-
ter giving consent, participants rated the humility of one of their
parents (75% chose to rate their mother) using the RHS–71
through a secure online medium. After completing question-
naires, participants were debriefed and given the contact infor-
mation of the researcher should they have any questions.
Results and Discussion
The correlation matrix for all humility items was analyzed us-
ing an EFA with maximum likelihood estimation (Fabrigar, We-
gener, MacCallum, & Strahan, 1999) with an oblique (OBLIM)
rotation. To determine the number of factors to extract, we con-
ducted a Scree test (Cattell, 1966) as well as a parallel analysis
(see Steger, 2006). The Scree test suggested a three-factor solu-
tion, whereas parallel analysis suggested a six-factor solution.
We decided to only retain the first three factors for sev e r a l r e a -
sons: (a) only the three factors had enough items to constitute a
reliable subscale; (b) items on the fourth factor were related to
moral character (e.g., he or she is a moral person), but not neces-
sarily humility; and (c) items on the fifth and sixth factor shared
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
228 DAVIS ET AL.
TABLE 1.—Factor loadings for Relational Humility Scale (RHS) in Study 1.
MSDGH S AVS
Global Humility
He/she has a humble character. 2.78 0.98 .90 −.04 .00
He or she is truly a humble person. 2.98 0.99 .81 −.07 −.02
Most people would consider
him/her a humble person.
2.76 1.06 .70 .02 .01
His or her close friends would
consider him/her humble.
3.01 0.92 .63 −.03 −.05
Even strangers would consider
him/her humble.
2.68 0.99 .62 .04 −.03
Superiority
He/she thinks of him/herself too
highly.
0.76 0.98 −.14 −.86 −.03
He/she has a big ego. 0.72 0.98 −.18 −.78 −.06
He/she thinks of him/herself as
overly important.
0.88 1.01 −.11 −.74 .00
Certain tasks are beneath him/her. 0.91 1.05 .06 −.67 .03
IfeelinferiorwhenIamwith
him/her.
0.98 1.11 .10 −.60 −.02
He/she strikes me as self-righteous. 1.16 1.23 .06 −.57 .05
He/she does not like doing menial
tasks for others.
0.98 1.09 −.01 −.52 .03
Accurate View of Self
He/she knows him/herself well. 2.85 0.94 .05 −.05 .87
He/she knows his/her strengths. 2.83 0.98 .02 .09 .74
He/she knows his/her weaknesses. 2.58 1.07 .02 −.07 .72
He/she is self-aware. 2.33 1.12 −.02 .00 .57
Eigenvalue 6.31 2.50 1.79
Variance accounted 36.99 12.88 9.11
Note. N =300. Values shown in bold loaded on primary factor;GH =Global Humility
subscale; S =Superiority subscale; AVS =Accurate View of Self subscale.
acommonword(e.g.,admire), but they did not align clearly
with a quality of humility, as defined in previous literature.
After examining the content of items, the factors were named
Global Humility (e.g., He/she has a humble character), Supe-
riority (e.g., He/she has a big ego), and Accurate View of Self
(e.g., He/she knows him/herself well). Items were dropped that
did not load at least .50 on their primary factor, or that loaded
over .25 on any secondary factor. In addition, two items were
dropped from the Global Humility factor because they were
highly redundant with other items.
The final version of the RHS consisted of 16 items, with three
factors that are theoretically consistent with our definition of
relational humility (Global Humility—5 items; Superiority—7
items; Accurate View of Self—4 items). Descriptive statistics
and factor loadings for RHS are listed in Table 1. The three
factors accounted for 63.9% of the variance in items. The Cron-
bach’s alphas for the full scale and subscales were .89 for the
full scale, .92 for Global Humility, .87 for Superiority, and .82
for Accurate View of Self. The intercorrelations among the sub-
scales were r=−.45 for Global Humility and Superiority; r=
.40 for Global Humility and Accurate View of Self; and r=
−.26 for Superiority and Accurate View of Self. These findings
provide initial evidence that the RHS has a three-factor struc-
ture. Because items were winnowed based on the characteristics
of one sample, we sought to confirm the factor structure in a
different sample in Study 2.
STUDY 2
Study 1 yielded a three-factor solution for a 16-item scale
of humility. The purposes of Study 2 were to (a) replicate the
three-factor structure of the RHS using a different sample and
(b) provide additional evidence of the estimated internal consis-
tency of the scale and subscales. Accordingly, we used CFA to
test the structure of the RHS.
Method
Participant s . Participants were 196 undergraduate students
(122 female) from a large urban university. Ages ranged from
18 to 42 (M=19.2, SD =2.95). The sample was ethnically
diverse (55.1% White/Caucasian, 16.8% Black/African Amer-
ican, 17.3% Asian/Asian American, 4.1% Latino/Latina, and
6.7% other or did not report).
Measure: Humility judgments (RHS). Humility was as-
sessed with the 16-item RHS, developed in Study 1. For this
sample, the Cronbach’s alpha coefficients were .90 for the Full
Scale, .92 for the Global Humility subscale, .82 for the Superi-
ority subscale, and .79 for the Accurate View of Self subscale.
The intercorrelations among the subscales were r=−.54 for
Global Humility and Superiority, r=.40 for Global Humil-
ity and Accurate View of Self, and −.34 for Superiority and
Accurate View of Self.
Procedure. Participants were recruited from undergradu-
ate classes and participated as part of a course requirement or
in exchange for partial course credit. After giving consent, par-
ticipants rated the humility of a parent (77% rated their mother)
using the RHS through a secure online medium. After com-
pleting questionnaires, participants were debriefed and given
the contact information of the researcher should they have any
questions.
Results and Discussion
The covariance matrix was analyzed with maximum likeli-
hood estimation using Mplus 8.1 (Muth´
en & Muth´
en, 2010).
Items of the RHS were used as indicators of the Global Hu-
mility, Superiority, and Accurate View of Self factors, which
were modeled as correlated factors. Several fit indexes were ex-
amined to evaluate the overall fit of the model—the chi-square
value, the comparative fit index (CFI), the square root mean
residual (SRMR), and the root mean square error of approxi-
mation (RMSEA). As a rule of thumb, a CFI around .95, an
SRMR equal to or less than .08, and an RMSEA equal to or
less than .06 suggest good fit (Hu & Bentler, 1999). We found
that the three-factor model showed good fit, χ2(101, N=196)
=168.21, p<.001, CFI =.96, SRMR =.05, RMSEA =.06.
We also compared the three-fa c t o r m o d e l t o a o n e - f a c t o r m o d e l
using a chi-square difference test. Fit worsened substantially for
the one-factor model, "χ2(3) =574.59, p<.001. Therefore,
the three-factor solution was retained. Thus, the results of Study
2provideadditionalevidenceforthethree-factorstructureand
internal consistency of the RHS.
STUDY 3
In Study 3, we manipulated humility experimentally. In a
between-subject design, we randomly assigned participants to
nominate the most or least humble person they have known. We
hypothesized that the RHS subscales would detect differences
between the two groups, even after controlling for characteristics
of the relationship between the participant and the target person.
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
RELATIONAL HUMILITY 229
Method
Participant s . Participants were 200 undergraduate students
(112 female) from a large urban university. Ages ranged from
18 to 39 (M=19.2, SD =2.9). The sample was ethnically
diverse (55.0% White/Caucasian, 17.0% Black/African Amer-
ican, 17.5% Asian/Asian American, 4.0% Latino/Latina, and
6.0% other or did not report).
Measures.
Humility Judgments: Humility was assessed using the
RHS. Cronbach’s alphas for the Full Scale and for the Global
Humility, Superiority, and Accurate View of Self subscales were
.95, .97, .90, and .90, respectively.
Similarity and Closeness: Similarity and closeness to the
target person were assessed using single-item measures (i.e.,
“how similar are you to him or her”; “how close do you feel
to him or her”). Items were completed on a 5-point rating scale
from 1 (not at all similar [close])to5(very similar [close]).
Positive and Negative Affect: Positive and negative affect
toward the target person were assessed using the 20-item Posi-
tive and Negative Affect Schedule (PANAS; Watson, Clark, &
Tellegen, 1988). Items were completed using a 5-point rating
scale from 1 (very slightly or not at all)to5(extremely). Watson
et al. (1988) reported Cronbach’s alphas ranging from .84 to .90
for the two subscales and temporal stability estimates ranging
from .39 to .71. Watson et al. also reported evidence of con-
struct validity. Scores on the scale were found to be related to
measures of general distress and dysfunction, depression, and
anxiety. For this sample, Cronbach’s alphas for the positive and
negative factors were .90 and .94, respectively.
State Empathy: Empathy toward the target person was
assessed using the 8-item Batson Empathy Adjectives (BEA;
Coke, Batson, & McDavis, 1978). Participants rated each ad-
jective (e.g., concerned) on a 6-point scale from 0 (not at all)
to 5 (extremely). Coke et al. (1978) reported Cronbach’s alphas
ranging from .79 to .95. Scores on the scale show evidence of
construct validity and were found to be positively correlated
with empathic concern, perspective taking, and helping behav-
ior. The Cronbach’s alpha for the current sample was .94.
Procedure. Participants were recruited from undergraduate
classes and participated in exchange for partial course credit.
After giving consent, participants completed an online survey.
Participants were alternately assigned to one of two conditions.
In the most humble condition, participants described their re-
lationship with a particular person they saw as most humble
(40% rated a parent, 37.4% a friend, 16.2% a relative other than
a parent, and 6% some other relationship). In the least hum-
ble condition, participants described their relationship with the
person they saw as least humble (8.9% rated a parent, 54.5%
a friend, 7.9% a relative other than a parent, and 28.7% some
other relationship). Participants assessed their relationship with
the target person, including humility, similarity, closeness, and
emotions toward the person (e.g., positive, negative, empathy).
After completing questionnaires, participants were debriefed
and given the contact information of the researcher should they
have any questions.
Results and Discussion
To test whether conditions wo u l d d i f f e r o n t h e R H S s u b -
scales, four univariate analyses of variance were conducted that
compared RHS total scores and subscale scores between the two
experimental conditions. To adjust for Type I error, a Bonferroni
correction was used. As predicted, Humility Judgments (RHS
Total) were higher in the most humble condition (M=68.96,
SD =6.44) than in the least humble condition (M=38.78,
SD =12.02), F(1, 194) =470.87, p<.001, d=3.13. For the
subscales, participants in the most humble condition (relative
to the least humble condition) reported higher scores on Global
Humility (M=22.17, SD =2.59 relative to M=9.26, SD =
4.89, p<.001, d=3.30); lower scores on Superiority (M=
17.03, SD =6.84 relative to M=30.07, SD =4.11, p<.001,
d=–2.31); and higher scores on Accurate View of Self (M=
16.65, SD =2.87 relative to M=12.48, SD =4.31, p<.001,
d=1.14).
We also tested whether the RHS detected diff e r e n c e s b e t w e e n
groups, after controlling for other relational characteristics (i.e.,
similarity, closeness, positive and negative emotion, and empa-
thy). A hierarchical multiple regression analysis was conducted,
with humility judgments as the criterion variable, relational
characteristics entered in Step 1, and experimental condition
entered in Step 2. Relational characteristics accounted for 70%
of the variance in RHS scores, F(5, 167) =78.59, p<.001. Con-
trolling for relational characteristics, condition still accounted
for an additional 10% of the variance in RHS scores, "R2=
.10, F(1, 166) =90.54, p<.001.
The results of Study 3 provided initial evidence supporting
the construct validity of the RHS. The RHS was sensitive to
participants’ nominations of people with high or low humility.
Even after controlling for other relational differences, the RHS
detected differences between the two conditions.
STUDY 4
The purpose of Study 4 was to provide evidence of the con-
struct validity for the RHS and to test one of the key propositions
of our model of relational humility. Namely, according to our
model of relational humility, conflict is one of the situations
in which humility is most observable. Conflict tends to cause
anger, contempt, disgust, and other negative emotions that gen-
erally evoke stubborn and arrogant behaviors. Thus, we wanted
to show that humility judgments would predict positive rela-
tionship characteristics in the context of conflict. We examined
humility in the context of conflict in two relationships.
First, we examined the participants’ relationship with some-
one who had hurt or offended them (i.e., called “the offender”).
Our model of relational humility suggests that humility should
be associated with other-oriented emotions toward the target.
Perceiving that a target person is humble involves seeing that
person as interpersonally other-oriented, lacking superiority, and
having an accurate view of self. Furthermore, we theorized that
humility is important in situations that challenge the hierarchy
of a relationship such as conflict. Thus, we hypothesized that
humility would be positively correlated with forgiveness and
empathy.
Second, we examined participants’ relationship with a par-
ent with whom they tend to have the most conflict. Gener-
ally, experiencing conflict in a relationship causes a judge to
develop increasingly negative views about the target person’s
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
230 DAVIS ET AL.
character (Baumeister et al., 2007). Thus, we hypothesized that
humility judgments would be related to current emotions toward
the parent and to perceptions of other virtues. Specifically, we
hypothesized that participants who perceived their parent to be
high in humility would report higher levels of positive emotion
and lower levels of negative emotion toward that parent. Also,
we hypothesized that participants who rated their parent higher
in humility would also rate their parent higher on other virtues.
Method
Participant s . Participants were 150 undergraduate students
(95 female) from a large urban university. Ages ranged from
18 to 28 (M=18.9, SD =1.5). The sample was ethnically
diverse (53.9% White/Caucasian, 17.1% Black/African Amer-
ican, 19.1% Asian/Asian American, 2.6% Latino/Latina, and
7.2% other or did not report).
Measures Related to an Offender.
Humility Judgments: Participants completed the RHS for
the offender. Cronbach’s alphas for the full scale and for the
Global Humility, Superiority, and Accurate View of Self sub-
scales were .89, .94, .85, and .89, respectively.
Unforgiveness: Unforgiveness toward the offender was as-
sessed with the 12-item Transgression Related Interpersonal
Motivations (TRIM; McCullough et al., 1998). The TRIM has
two subscales: Revenge (5 items; e.g., “I’ll make him/her pay”)
and Avoidance (7 Items; e.g., “I keep as much distance between
us as possible”). Items were completed on a 5-point rating scale
from 1 (strongly disagree)to5(strongly agree)Lowerscores
indicate higher forgiveness (i.e., lower unforgiveness). McCul-
lough et al. (1998) reported Cronbach’s alpha coefficients for
scores on both subscales ranging from .85 to .93 and estimates
of 3-week temporal stability ranging from .44 to .65. Scores on
the scale showed evidence of construct validity and were found
to be negatively correlated with other measures of forgiveness,
relationship satisfaction, and commitment. In this sample, Cron-
bach’s alphas were .90 for Revenge and .94 for Avoidance.
State Empathy: Empathy toward the target offender was
assessed with the 8-item BEA (Coke et al., 1978; described in
Study 3). In the current sample, the Cronbach’s alpha was .94.
Measures Related to a Parent.
Humility Judgments: Participants completed the RHS for
the parent with whom they have had the most conflict. Cron-
bach’s alphas for the full scale and for the Global Humility,
Superiority, and Accurate View of Self subscales were .92, .97,
.85, and .90, respectively.
Positive and Negative Affect in Rela t i o n s h i p w i th Pare n t :
Positive and negative affect toward the parent were assessed with
the 20-item PANAS (Watson et al., 1988; described in Study 3).
In this sample, Cronbach’s alphas for Positive Emotions and
Negative Emotions were .88 and .92, respectively.
Virtu e Judgments: The degree to which parents were
viewed as possessing various virtues was assessed with the
18-Item Virtue Orientation Scale (VOS; Berry, Worthington,
O’Connor, Parrott, & Wade, 2004; Berry, Worthington, Wade,
Witvliet, & Kiefer, 2005). Items were rated with a 10-point rat-
ing scale ranging from 1 (not at all)to10(very much). The
VOS has two subscales: warmth-based virtues (e.g., empathy or
forgiveness) and conscientiousness-based virtues (e.g., justice
or self-control). The psychometric properties of this scale have
been reported in previous research (Berry et al., 2004; Berry et
al., 2005). Cronbach’s alpha coefficients have ranged from .71
to .94. Furthermore, the items of each scale have demonstrated
adequate fit to a Rasch model for rating scales, demonstrating
that the items are forming adequate unidimensional scales. In
this sample, Cronbach’s alphas for Warmth-Based Virtues and
Conscientious-Based Virtues were .94 and .95, respectively.
Procedure. Participants were recruited from undergraduate
classes and participated for partial course credit. After giving
consent, participants first thought about a person who had hurt
or offended them, wrote a summary of the transgression, and
completed a set of questionnaires about their relationship with
the offender. Second, participants thought about the parent with
whom they experienced the highest level of conflict (62.6%
chose their mother), wrote a description of this relationship,
and completed a second set of questionnaires about their rela-
tionship with the target parent. After completing questionnaires,
participants were debriefed and given the contact information
of the researcher should they have any questions.
Results and Discussion
Means, standard deviations, alphas, and intercorrelations for
all scales are reported in Table 2. To show evidence of construct
validity, we examined the correlations of the RHS with other
relationship characteristics.
First, we hypothesized that perceived humility of an offender
would be negatively related to unforgiveness and positively re-
lated to empathy toward that offender. This hypothesis was sup-
ported. The total RHS, Global Humility, and Accurate View of
Self subscale scores were related to less total unforgiveness,
as well as avoidance and revenge, and were related to greater
empathy; likewise, the Superiority subscale was related to more
total unforgiveness, as well as avoidance and revenge, and was
related to less empathy (see Tabl 2).
Second, we hypothesized that humility judgments of a par-
ent would be positively related to perceptions of virtue and
current experiences of emotion toward that parent. This hy-
pothesis was supported. Higher levels of perceived humility
were related to higher perceptions of both warmth-based and
conscientiousness-based virtues (see Table 2). Higher levels of
perceived humility were also related to higher positive emo-
tion and lower negative emotion toward the target parent (see
Table 2).
In Study 4, we provided additional evidence supporting the
construct validity of the RHS. Greater levels of perceived humil-
ity were related to (a) greater forgiveness of and empathy toward
an offender, and (b) greater perceptions of warmth-based and
conscientious-based virtues, greater positive emotions, and less
negative emotions toward a target parent.
STUDY 5
The purposes of Study 5 were to (a) provide additional ev-
idence that humility judgments are related to the judge’s rela-
tionship with an offender, and (b) to examine evidence of the
discriminant and incremental validity of the RHS compared with
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
RELATIONAL HUMILITY 231
TABLE 2.——Intercorrelations of the Relational Humility Scale (RHS) with other constructs (Study 4).
Tot a l S ubs c a les
aMSDRHS GH S AVS
Relationship with offender
Unforgiveness .94 29.63 12.63 −.49∗∗ −.45∗∗ .33∗∗ −.37∗∗
Avoidance .90 19.77 9.01 −.50∗∗ −.43∗∗ .32∗∗ −.40∗∗
Revenge .94 9.87 5.14 −.35∗∗ −.35∗∗ .25∗∗ −.22∗∗
Empathy .94 19.60 8.16 .49∗∗ .48∗∗ −.30∗∗ .37∗∗
Relationship with parent
Warmth virtues .94 55.12 20.31 .63∗∗ .59∗∗ −.48∗∗ .49∗∗
Conscientiousness virtues .95 49.09 18.48 .60∗∗ .56∗∗ −.46∗.50∗∗
Positive emotions .88 30.64 9.30 .57∗∗ .54∗∗ −.41∗∗ .54∗∗
Negative emotions .92 22.07 7.80 −.41∗∗ −.25∗∗ .44∗∗ −.20∗
Note. N =142. GH =Global Humility subscale; S =Superiority subscale; AVS =Accurate View of Self.
∗p<.05; ∗∗p<.01.
the HH subscale of the HEXACO–PI (Lee & Ashton, 2004). In
this study, we recruited participants who had been hurt by a
father’s dedication to his work. This offense aligns with two of
the three situations in which humility should be observable (i.e.,
conflict in a relationship with an authority figure). We predicted
that judgments of humility—measured with both the RHS and
the HH—would affect the extent to which participants had ex-
perienced forgiveness toward their father.
Method
Participant s . Participants were 163 undergraduate students
(95 female) from a large urban university. Ages ranged from
18 to 25 (M=20.0, SD =2.6). The sample was ethnically
diverse (54.0% White/Caucasian, 20.2% Black/African Amer-
ican, 9.2% Asian/Asian American, 3.1% Latino/Latina, and
13.5% other or did not report).
Measures.
Humility Judgments: Participants completed the RHS to
describe the humility of the target person. Cronbach’s alphas for
Global Humility, Superiority, and Accurate View of Self were
.93, .89, and .89, respectively. They also completed the 16-item
HH subscale of the HEXACO–PI–100 (Lee & Ashton, 2004)
for the target. Participants rated their father on each item (e.g.,
“He/she is an ordinary person who is no better than others”) on
a 5-point scale from 1 (strongly disagree)to5(strongly agree)
as applied to the target. The HH has four facet scales: Sincerity,
Fairness, Greed-Avoidance, and Modesty. The Cronbach’s alpha
for the full scale was .81; the facet scales ranged from .47 to .74
(see Table 3). Thus, the alpha was adequate for the full scale
score, but several of the facet scores had low reliability. The HH
subscale has considerable evidence of incremental validity, over
and above the five-factor model (Ashton & Lee, 2007).
Unforgiveness: Unforgiveness was assessed using the 12-
item TRIM (McCullough et al., 1998; see description in Study
4). Cronbach’s alphas for the avoidance and revenge subscales
were .95 and .87, respectively.
Procedure. Participants were recruited from undergraduate
classes and participated for partial course credit. After giving
consent, participants thought of a time when they had been
hurt by their father’s devotion to work. They then completed
measures regarding the offense and their relationship with their
father. After completing questionnaires, participants were de-
briefed and given the contact information of the researcher
should they have any questions.
Results and Discussion
Means, standard deviations, alphas, and intercorrelations for
all scales are reported in Table 3. First, we hypothesized that
the RHS subscales would show evidence of discriminant valid-
ity from the HH subscales. The Global Humility subscale was
moderately correlated with HH; the Superiority subscale was
strongly (and negatively) related; and the Accurate View of Self
was weakly related (see Table 3).
Second, we hypothesized that the RHS would predict unfor-
giving motivations above and beyond the HH subscales. To test
this hypothesis, we conducted two hierarchical regression anal-
yses with avoidance or revenge motivations as the dependent
variables. For each analysis, the HH subscales were entered in
the first step and the RHS subscales were entered into the second
step. The HH subscales predicted 32% of the variance in revenge
scores in the first step, and the RHS predicted an additional 8%
of the variance in revenge scores in the second step, R2"=.08,
F(3, 140) =4.63, p=.004. The HH subscales predicted 43%
of the variance in avoidance scores in the first step, and the RHS
predicted an additional 5% of the variance in avoidance scores
in the second step, R2"=.05, F(3, 138) =3.29, p=.022.
We also analyzed the rev e r s e m o d e l s , examining whether
the HH predicted unforgiving motivations after controlling for
the subscales of the RHS. The RHS subscales predicted 41% of
the variance in revenge scores in the first step, and the HH did not
predict significant additional variance in revenge scores in the
second step, R2"=.01, F(1, 143) =2.05, p=.154. However,
the RHS subscales predicted 38% of the variance in avoidance
scores in the first step, and the HH predicted an additional 7%
of the variance in avoidance scores in the second step, R2"=
.07, F(1, 141) =11.71, p=.001.
We conducted ancillary analyses to determine whether a
brief version of the RHS—the Global Humility subscale
alone—would also explain additional variance in revenge and
avoidance motivations, above and beyond the HH subscales. If
it did so, some researchers might wish to use this brief version of
the scale. Thus, we conducted two hierarchical regression anal-
yses with avoidance or revenge motivations as the dependent
variable, the HH entered in a first step, and the Global Humility
subscale entered in a second step. The Global Humility subscale
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
232 DAVIS ET AL.
TABLE 3.—Intercorrelations of the Relational Humility Scale (RHS) with the HH and forgiveness (Study 5).
MSDNαRHS GH S AVS 4 5 6 7 8 9
1. RHS Total 59.04 11.73 154 .91 —
2. RHS Global Humility 17.48 5.78 160 .93 .83∗∗
3. RHS Superiority 12.15 8.90 160 .89 −.83∗∗ −.49∗∗
4. RHS Accurate View of Self 14.61 4.48 160 .89 .62∗∗ .43∗∗ −.21∗∗
5. HH Sincerity 12.50 3.28 161 .47 .27∗∗ .16∗−.27∗∗ .10
6. HH Fairness 15.23 3.43 163 .61 .41∗∗ .31∗∗ −.39∗∗ .21∗∗ .33∗∗
7. HH Greed Avoidance 11.90 4.07 162 .74 .38∗∗ .27∗∗ −.37∗∗ .10 .42∗∗ .24∗∗
8. HH Modesty 13.65 3.58 162 .66 .60∗∗ .47∗∗ −.53∗∗ .33∗∗ .39∗∗ .44∗∗ .58∗∗
9. HH Full Scale Score 53.33 10.64 159 .81 .56∗∗ .42∗∗ −.53∗∗ .25∗∗ .71∗∗ .66∗∗ .78∗∗ .82∗∗
10. TRIM Avoidance 12.67 6.98 159 .87 .37∗∗ −.31∗∗ .21∗∗ −.25∗∗ −.36∗∗ −.27∗∗ −.24∗∗ −.29∗∗ −.40∗∗
11. TRIM Revenge 6.49 2.87 161 .95 .40∗∗ −.29∗∗ .35∗∗ −.22∗∗ −.21∗∗ −.22∗∗ −.22∗∗ −.27∗∗ −.32∗∗ .67∗∗
Note. RHS =Relational Humility Scale; HH =Honesty-Humility; TRIM =Transgression-related interpersonal motivations; GH =Global Humility subscale; S =Superiority
subscale; AVS =Accurate View of Self.
∗p<.05; ∗∗p<.01.
predicted an additional 3% of the variance in revenge scores,
R2"=.029, F(1, 148) =4.97, p=.027, and an additional 3.7%
of the variance in avoidance scores, R2"=.037, F(1, 146) =
6.82, p=.01, over and above the HH.
Thus, in Study 5, we provided additional evidence for the
model of relational humility, finding that both measures of re-
lational humility (i.e., the RHS and the HH) were related to the
degree to which participants had forgiven an offense related to
afather’sdevotiontowork.Wealsoshowedevidenceofdis-
criminant validity for the RHS, showing that the RHS subscales
differed from the HH and its subscales. Furthermore, measur-
ing humility judgments with the RHS incrementally predicted
forgiveness, above and beyond the HH, and we also provided
evidence that a briefer version of the RHS—administering only
the Global Humility scale—explained unique variance in unfor-
giving motivations.
GENERAL DISCUSSION
The main purpose of this research was to describe a model
of relational humility and to provide evidence for that model.
Second, we developed a face-valid measure of humility judg-
ments, which allows continued progress in measuring humility
through other-reports. Thus, in a series of five studies, we de-
veloped the 16-item RHS (Study 1). We replicated its factor
structure (Study 2), which provided construct and content valid-
ity for the construct of relational humility. We demonstrated the
sensitivity of the RHS to rating people nominated as high or low
in humility, even after controlling for relational characteristics
(Study 3). We also provided evidence supporting the construct
validity of the RHS scale (Study 4). Finally, we provided evi-
dence that the RHS subscales account for unique variance, above
and beyond the 16-item HH subscale of the HEXACO–PI–100
(Lee & Ashton, 2004; Study 5). In psychological science, defi-
nition, concept, and measurement are intertwined. In these five
studies, we provided evidence for the way of defining, concep-
tualizing, and measuring relational humility, as described by our
theoretical model.
The RHS was found to have three subscales—Global Humil-
ity, Superiority, and Accurate View of Self—that correspond
with our definition of relational humility. The factor structure
of the RHS was replicated in an independent sample, and the
subscales showed acceptable estimates of internal consistency.
Furthermore, the RHS and its subscales showed initial evidence
of construct validity. The subscales were found to be correlated
with other positive relationship characteristics and dynamics,
such as forgiveness, empathy and other emotions, and percep-
tion of virtues.
Studies 4 and 5 were the first to focus on humility of an
offender in the context of forgiveness. Previous theory and re-
search has focused on humility of the victim, proposing that
humility can promote empathy and dissipate feelings of dis-
gust and moral superiority toward the offender (e.g., Sandage,
1999; Worthington, 1998). For example, self-reports of humility
have been positively associated with trait forgivingness (Rowatt
et al., 2006; Shepherd & Belicki, 2008). Moreover, victims have
been found to be more likely to forgive when they considered
themselves capable of committing a similar offense (Exline,
Baumeister, Zell, Kraft, & Witvliet, 2008; Exline, Zell, Mal-
colm, DeCourville, & Belicki, 2008). Whereas these studies
focused on humility in the victim, we focused on the degree to
which the victim judged the offender as possessing humility.
In both Studies 4 and 5, perceiving an offender as humble was
positively related with greater forgiveness.
With the articulation of a model of relational humility, as
well as the development of the RHS, researchers have a choice
of measures to begin studying relational humility. An advantage
of the RHS is that its items and subscales were evaluated and de-
veloped using both theory and empirical methods. Its subscales
align with how humility has been defined (Davis, Worthington,
et al., 2010). In addition, the RHS is the only measure that has
been developed primarily to assess the degree an observer judges
atargetpersontopossesshumility.Furthermore,themeasureis
brief. Indeed, the Global Humility subscale (5 items) provides a
brief snapshot of relational humility (in Study 5, we found that
the 5-item measure predicted unique variance in avoidance and
revenge motivations, above and beyond the HH—albeit only a
small amount of additional variance was accounted for). The
other two subscales (11 items) provide a more nuanced assess-
ment of humility.
Limitations
This set of studies had several limitations. First, only one
method of measurement was used (i.e., other-report). In the fu-
ture, it will be important to examine consensus among raters and
agreement with self-report. In addition, researchers should ex-
amine how humility judgments are related to physiological (e.g.,
are RHS ratings related to more emotional calmness during, for
example, discussions of relational trust?), behavioral (e.g., are
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
RELATIONAL HUMILITY 233
RHS ratings related to fewer uses of personal pronouns, fewer
defensive verbalizations, and fewer nonverbal defensive behav-
iors?), or cognitive (e.g., are RHS ratings related to scores on
the IAT–HA or some other test of cognition) measures. Second,
this set of studies used cross-sectional rather than longitudinal
designs, and thus we were unable to assess how humility judg-
ments might change over time. For example, humility judgments
might be sensitive to conflict in a relationship. Similarly, high
fluctuations in relational humility judgments might be associ-
ated with negative relational qualities, such as low commitment
or satisfaction. Third, only college students were studied (see
Henrich, Heine, & Norenzayan, 2010, who critique the sole use
of undergraduates). Thus, it is important to evaluate the RHS,
as well as the model of relational humility, in other samples, es-
pecially community samples in which participants are involved
in hierarchical roles (e.g., mentorship, supervisor–subordinate,
or teacher–student relationships).
Future Research
With the development of the RHS, researchers now have at
least two strong options for measuring relational humility, in-
cluding the RHS and the HH. Given that humility research is
in an early phase, we suspect that other researchers will put
forth alternative definitions of humility and develop associated
measures. To provide assurance in measurement, it is neces-
sary to continue to evaluate the construct and criterion-related
validity of the RHS, as well as other measures. Researchers
should explore the unique correlates of the three subscales. For
example, the Accurate View of Self subscale might uniquely
predict self-monitoring, whereas the Superiority subscale might
uniquely predict Machiavellianism. Furthermore, it is important
to explore some of the constructs, behaviors, and physiological
markers that are associated with judgments of high humility. We
have theorized that restraining expressions of low humility, as
measured by the Superiority subscale, might have different cor-
relates than expressing high humility, as measured by the Global
Humility and Accurate View of Self subscales. However, it is
also possible that the factor structure of the RHS reflects whether
items were positively or negatively keyed.
Now that initial measures of relational humility exist, re-
searchers should turn to testing some of the many implications
of our theoretical model. First, one might be able to derive an
estimate of trait humility using consensus among several in-
formants (e.g., Davis, Worthington, et al., 2010; Kenny, 2004).
However, if consensus between raters is particularly low, sug-
gesting that ratings of humility are relationship-dependent, it
might not make sense to derive a trait estimate.
Second, researchers might explore the self–other agreement
of humility ratings. For instance, this method could be used to
actually examine the degree to which humble people (as rated
by others) modestly report their own humility. Researchers have
yet to provide evidence for this hypothesis, which is assumed to
undermine the validity of self-reports of humility.
Third, RHS scores could also be treated as a criterion variable
to offer evidence of validity for other measures, such as the
IAT–HA (Rowatt et al., 2006). Showing that the IAT–HA is
strongly related to the RHS subscales would provide important
evidence that the measure predicts whether people are actually
seen as humble across their relationships.
Fourth, researchers should examine our claim that humility
is most readily observed in three situations: receiving praise
or recognition, hierarchical roles, or conflict. For example, re-
searchers might experimentally manipulate these factors that
make humility more difficult to practice and compare how self-
reports and other-reports predict differences in behavior across
conditions. Similarly, researchers should begin to study humil-
ity in samples for which humility is potentially important, such
as business leaders, teachers, elected public officials, parents, or
couples.
Fifth, researchers should examine evidence for several nu-
anced hypotheses implied by our theoretical model. For ex-
ample, our model suggests that humility is subjectively eas-
ier (and less ambiguous) to assess in others than in oneself.
Modesty norms are likely more salient when describing one’s
own humility than when judging someone else’s humility. Also,
defensive processes of maintaining a particular conceptualiza-
tion of oneself could be lessened. Moreover, researchers might
examine whether relationship factors moderate how humility
is attributed within relationships (see Funder, 1995, for a re-
view). For example, convergence between self- and observer-
reports should be higher in relationships between people who
know each other well than in relationships between strangers
(Kenny, 2004). Also, people might consider different qualities
when judging the humility of close friends versus acquaintances
(McAdams & Pals, 2006). Furthermore, cultural or spiritual fac-
tors might affect how people judge humility (Davis, Hook, et
al., 2010; Emmons & Kneezel, 2005; Sandage, 1999). Finally,
we hypothesize that humility should be less stable over time
than other traits such as extraversion. For example, only a few
immodest acts could squander a reputation of humility.
CONCLUSION
For some time, humility researchers have been unsatisfied
with measures of humility (e.g., Tangney, 2009). Whereas mod-
esty researchers have been satisfied with separately studying
different aspects of modesty, such as accurate view of self or
modest self-presentation, humility researchers have insisted on
defining humility as integrity among thought, action, and mo-
tivation. Such noble conceptualizations of humility have made
humility difficult to assess. Only a few measures of humility
have been published in peer-reviewed journals (e.g., Lee & Ash-
ton, 2004; Rowatt et al., 2006). We have proposed a model of
relational humility that contextualizes humility within actual
relationships and studies humility from the perspective of a
third-party observer. The approach aligns the study of humility
with an expansive body of research on personality judgments
and circumvents the paradox of using self-reports of humility.
Furthermore, we developed a scale of humility that aligns with
our proposed definition and approach for studying humility. We
hope the introduction of this model and scale will be an impor-
tant step in catalyzing research on humility.
ACKNOWLEDGMENT
Gratitude is expressed to the John Templeton Foundation for
funding the current project.
REFERENCES
Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and practical advan-
tages of the HEXACO model of personality structure. Perso n a lit y a n d S oc i a l
Psychology Review,11, 150–166.
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
234 DAVIS ET AL.
Baumeister, R. F., Vohs, K. D., DeWall, C. N., & Zhang, L. (2007). How
emotion shapes behavior: Feedback, anticipation, and reflection, rather than
direct causation. Personality and Social Psychology Review,11, 167–203.
Berry, J. W., Worthington, E. L., Jr., O’Connor, L., Parrott, L., III, & Wade,
N. G. (2004). Forgiveness, vengeful rumination, and affective traits. Journal
of Personality,73, 1–43.
Berry, J. W., Worthington, E. L., Jr., Wade, N. G., Witvliet, C. v. O., & Kiefer,
R. (2005). Forgiveness, moral identity, and perceived justice in crime victims
and their supporters. Humboldt Journal of Social Relations,29, 136–162.
Carver, C. S., & Scheier, M. F. (1994). Optimism and health-related cognition:
What variables actually matter? Psychology and Health,9, 191–195.
Cattell, R. B. (1966). The Scree test for the number of factors. Multivariate
Behavioral Research,1, 245–276.
Coke, J. S., Batson, C. D., & McDavis, K. (1978). Empathic mediation of
helping: A two-stage model. Journa l of Personality and Social Psychology,
36, 752–766.
Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality Inventory
(NEO–PI–R) professional manual.Odessa,FL:PsychologicalAssessment
Resources.
Davis, D. E., Hook, J. N., Worthington, E. L., Jr., Van Tongeren, D. R., Gartner,
A. L., & Jennings, D. J., II. (2010). Relational spirituality and forgiveness:
Development of the Spiritual Humility Scale (SHS). Journal of Psychology
and Theology,38, 91–100.
Davis, D. E., Worthington, E. L., Jr., & Hook, J. N. (2010). Relational humil-
ity: A review of definitions and measurement strategies. Journal of Positive
Psychology,5, 243–252.
de Vries, R. E., Lee, K., & Ashton, M. C. (2008). The Dutch HEXACO Person-
ality Inventory: Psychometric properties, self–other agreement, and relations
with psychopathy among low and high acquaintanceship dyads. Journal of
Per s ona l i ty As s e ssm e n t,90, 142–151.
Emmons, R. A., & Kneezel, T. T. (2005). Giving thanks: Spiritual and religious
correlates of gratitude. Journal of Psychology and Christianity,24,140–148.
Emmons, R. A., & McCullough, M. E. (Eds.). (2004). The psychology of grati-
tude. New York, NY: Oxford University Press.
Exline, J. J., Baumeister, R. F., Zell, A. L., Kraft, A. J., & Witvliet, C. V. O.
(2008). Not so innocent: Does seeing one’s own capability for wrongdo-
ing predict forgiveness? Journal of Personality and Social Psychology,94,
495–515.
Exline, J. J., Zell, A. L., Malcolm, W., DeCourville, N., & Belicki, K. (2008).
Does a humble attitude promote forgiveness? Challenges, caveats, and sex
differences. In Women’s reflections on the complexities of forgiveness (pp.
235–251). New York, NY: Routledge/Taylor & Francis Group.
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C. & Strahan, E. J. (1999).
Evaluating the use of exploratory factor analysis in psychological research.
Psychological Methods,4, 272–299.
Fredrickson, B. L. (1998). What good are positive emotions? Review of General
Psychology,2, 300–319.
Funder, D. C. (1995). On the accuracy of personality judgment: A realistic
approach. Psychological Review,102, 652–670.
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: Attitudes,
self-esteem, and stereotypes. Psychological Review,102, 4–27.
Gregg, A. P., Hart, C. M., Sedikides, C., & Kumashiro, M. (2008). Every-
day conceptions of modesty: A prototype analysis. Per s ona l i ty an d S o c ia l
Psychology Bulletin,34, 978–992.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the
world? Behavioral and Brain Sciences,33, 61–83.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance
structure analysis: Conventional criteria versus new alternatives. Structural
Equation Modeling,6, 1–55.
John, O. P., & Robins, R. W. (1993). Determinants of interjudge agreement on
personality traits: The Big Five 820 domains, observability, evaluativeness,
and the unique perspective of the self. Journal of Personal ity,61, 521–551.
Kenny, D. A. (2004). PERSON: A general model of interpersonal perception.
Personality and Social Psychology Review,8, 265–280.
Kwan, V. S. Y., John, O. P., Kenny, D. A., Bond, M. H., & Robins, R. W.
(2004). Reconceptualizing individual differences in self-enhancement bias:
An interpersonal approach. Psychological Review,111, 94–110.
Kwan, V. S. Y., John, O. P., Robins, R. W., & Kuang, L. L. (2008). Conceptu-
alizing and assessing self-enhancement 835 bias: A componential approach.
Journal of Personality and So cial P sychology ,94, 1062–1077.
Lee, K., & Ashton, M. C. (2004). Psychometric properties of the HEXACO
personality inventory. Multivariate Behavioral Research,39, 329–358.
Lopez, S. J., & Snyder, C. R. (Eds.). (2009). Oxford handbook of positive
psychology. New York, NY: Oxford University Press.
McAdams, D. P., & Pals, J. L. (2006). A new Big Five: Fundamental principles
for an integrative science of personality. American Psychologist,61,204–217.
McCrae, R. R., & Weiss, A. (2007). Observer ratings of personality. In R. W.
Robins, R. C. Fraley, & R. F. Krueger (Eds.), Handbook of research methods
in personality psychology (pp. 259–272). New York, NY: Guilford.
McCullough, M. E., Rachal, K. C., Sandage, S. J., Worthington,E. L., Jr., Brown,
S. W., & Hight, T. L. (1998). Interpersonal forgiving in close relationships: II.
Theoretical elaboration and measurement. Journal of Person alit y and So cial
Psychology,75, 1586–1603.
Muth´
en, L. K., & Muth´
en, B. O. (2010). Mplus user’s guide (6th ed.). Los
Angeles, CA: Muth´
en & Muth´
en.
Park, N., & Peterson, C. (2006). Moral competence and character strengths
among adolescents: The development and validation of the Values in Action
Inventory of Strengths for Youth. Journal of Adolescence,29, 891–909.
Powers, C., Nam, R. K., Rowatt, W. C., & Hill, P. C. (2007). Associations
between humility, spiritual transcendence, and forgiveness. Research in the
Social Scientific Study of Religion,18, 75–94.
Rowatt, W. C., Ottenbreit, A., Nesselroade, K. P., Jr., & Cunningham,
P. A. (2002). On being holier-than-thou or humbler-than-thee: A social-
psychological perspective on religiousness and humility. Journa l for th e Sci-
entific Study of Religion,41, 227–237.
Rowatt, W. C., Powers, C., Targhetta, V., Comer, J., Kennedy, S., & Labouff, J.
(2006). Development and initial validation of an implicit measure of humility
relative to arrogance. Journal of Positive Psychology,1, 198–211.
Sandage, S. J. (1999). An ego-humility model of forgiveness: A theory-driven
empirical test of group interventions. Unpublished doctoral dissertation, Vir-
ginia Commonwealth University, Richmond, VA.
Sedikides, C., Gregg, A. P., & Hart, C. M. (2007). The importance of being
modest. In C. Sedikides & S. J. Spencer (Eds.), The self (pp. 163–184). New
York, NY: Psychology Press.
Shepherd, S., & Belicki, K. (2008). Trait forgiveness and traitedness within the
HEXACO model of personality. Personality and Individual Differences,45,
389–394.
Snyder, C. R., Rand, K. L., & Sigmon, D. R. (2002). Hope theory: A member
of the positive psychology family. In C. R. Snyder & S. J. Lopez (Eds.),
Handbook of positive psychology (pp. 411–419). New York, NY: Oxford
University Press.
Steger, M. F. (2006). An illustration of issues in factor extraction and identifica-
tion of dimensionality in psychological assessment data. Journal of Person -
ality Assessment,86, 263–272.
Tangney, J. P. (2005). Humility. In C. R. Snyder & S. J. Lopez (Eds.), Handbook
of positive psychology (pp. 411–419). New York, NY: Oxford University
Press.
Tan g n ey , J . P. (20 0 9 ). Hu m i lit y . I n S. J. L o p ez & C. R . S nyd e r ( Eds . ) , Oxford
handbook of positive psychology (pp. 483–490). New York, NY: Oxford
University Press.
Tangney, J. P., Stuewig, J., & Mashek, D. J. (2007). Moral emotions and moral
behavior. Annual Review of Psychology,58, 345–372.
Tice, D. M., But ler , J. L. , Mur aven, M. B., & Stillwell, A. M. (19 95) . Whe n
modesty prevails: Differential favorability of self-presentation to friends and
strangers. Journal of Personality and So cial P sychology ,69, 1120–1138.
Vazire, S. (2006). Informant reports: A cheap, fast, and easy method for person-
ality assessment. Journal of Research in Personality,40, 472–481.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation
of brief measures of positive and negative affect: The PANAS scales. Journal
of Personality and Social Psychology,54, 1063–1070.
Worthington, E. L. (1998). An empathy-humility-commitment model of for-
giveness applied within family dyads. Journal of Family Therapy,20,59–76.
Wor t h ing t o n, E. L . , J r. (E d . ). (2 0 0 5). Handbook of forgiveness. New York, NY:
BrunnerRoutledge.
Downloaded By: [Virginia Commonwealth University] At: 11:45 8 June 2011
